The Intelligent Texture Anti-counterfeiting Algorithm Based on DFT

This paper presents an intelligent texture anti-counterfeiting method based on human visual characteristics and DFT. By using a part of sign sequence of DFT coefficients as the feature vector of texture images, we extract a visual feature vector and then create a database of texture images´ feature vectors. After comparing the Normalized Cross-correlation (NC) between the feature vectors, we achieve the purpose of automatic identification. The experimental results show that the proposed algorithm has strong robustness against common and geometric attacks.